﻿*****************************************************************************
Paper: "Making comprehensible speeches when your constituents need it"
Journal: Research & Politics
Authors: Nick Lin (University of Mannheim) and Moritz Osnabrügge (Bocconi University)
Date: July 20, 2018
*****************************************************************************


Replication files:
1. Data ("Lin_Osnabruegge_RAP_data.dta")
2. Code to replicate the results and further analysis ("Lin_Osnabruegge_RAP_dofile.do")



*****************************************************************************


The data includes the following variables:

- name: name of speakers
- speakerid: ID number of speakers
- term: legislative term
- date: date of speech
- constituent_number: ID number of constituencies
- elected_constituent: whether a MP was elected by district elections
- party: MP's party (source: Rauh 2015)
- scorefre: FRE score (calculated using koRpus R package)
- scorelixr: Lix score, inverted (calculated using koRpus R package)
- unemployment: unemployment rate in a district (source: German statistical office)
- hochschulreife: share of students that leave school with a high school degree (source: German statistical office)
- nongermanpop: share of citizens without a German citizenship (source: German statistical office)
- seniority: number of years a politician served as MP in the Bundestag (source: http://zhsf.gesis.org/ParlamentarierPortal/index.htm)
- party leader (source: party websites or e-mail correspondence with party)
- high job: MP was an engineer, chemist, physicist, mathematician, entrepeneur, consultant, parliamentarian, doctor, social scientist (source: German statistical office)
- phd: MP has a Ph.D. degree (source: German Bundestag)
- age: MP's age (source: German statistical office)
- female: indicating female MPs
- Migration background: MP has a migration background (source: Blätte and Wüst 2017)
- speech length: number of words in a speech